A Multi-dimensional Vector Parallel Search Algorithms

To solve the quick search problem of multi-dimensional vector data, this paper constructed a multi-branch extended and balanced index tree, by using the nearest neighbor criterion to realize the space division of the data sets, under the condition that the upper limit of the query scope is known. Then, the serial and parallel search algorithms of the index tree were designed. At last, the capability of the parallel search algorithms was analysed and the test results validated the validity of this method.